Hydraulic flow and downstream scouring behavior on spillway system of Surumana Dam

Surumana Dam is designed to meet various needs of water in Donggala, Central Sulawesi, and its surroundings. It is planned to be equipped with a side channel spillway, which sequentially consists of an approach channel, side spillway, side channel, transition channel, slide channel, energy absorber type USBR II, and an escape channel to the main river. The safety of the spillway system design, which affects the overall safety of the dam, is determined by the hydraulic behavior. This study aims to investigate the hydraulics flow in the side spillway system and scouring in downstream due to the flood discharge through the spillway dam by using a physical model test in the laboratory with a scale of 1:60. The model was used to measure the hydraulic parameter as velocity, depth, and pressure, under Q100, Q1000 and QPMF discharge. The result of the analysis showed the hydraulic flow behavior, especially in the phenomenon of damping the energy of water flow in stilling ponds (energy damper) and scouring in the downstream river bed. The results of this hydraulic model test provided a better picture of the flow behavior of the spillway system in terms of energy dissipation and scour.


Introduction
Dam construction is an effort necessary to encourage economic growth and community welfare through the functions and benefits of the dam [1].Regarding this matter, Donggala Regency, Central Sulawesi Province, is an area with fairly dry climate conditions, so the Surumana Dam was built, which is located at upstream of the Surumana River.This dam provides irrigation water covering an area of 2,100 ha, ponds covering an area of 500 ha, domestic raw water of 1.20 m 3 /second, industry of 2.30 m 3 /second, and electricity generation of 1.5 MW.The total storage volume of the Surumana Dam is 45.14 million m 3 .This reservoir is intended as a reservoir for water runoff or floods and to provide water needs during the dry season so that it can improve community welfare and support the economic growth of the people of Donggala Regency.
In accordance with dam planning principles, spillway planning is an important part of dam construction, requiring in-depth study and accuracy [2].In planning spillway construction, it is necessary to know the behavior of the hydraulic flow conditions by carrying out hydraulic model tests [3].This was done to improve the dam spillway design and determine the actual flow conditions on the Surumana Dam spillway because hydraulic flow conditions on the spillway are difficult to approach using theoretical calculations alone.
This research aims to determine the suitability of hydraulic behavior when testing spillway hydraulic models starting from the upstream direction channel (approach channel), spillway, side channel, transition channel, launch channel, energy reducer, escape channel, and river on the behavior of flow hydraulic conditions with results theoretical calculation on the Q100, Q1000, and QPMF discharge with a discharge of 530.70 m 3 /sec.The different discharges will lead to different velocities.The increase in bed shear velocity results in scouring [4].Apart from that, from the implementation of spillway hydraulic model tests, the possibility of cavitation and local scour phenomena can be seen downstream of the Surumana Dam.This is done to improve the desired goal, namely to fulfill dam safety requirements from a hydraulic perspective and to make recommendations for change to get good flow hydraulic planning results in the spillway system [5].This research investigates the hydraulic behavior from the upstream spillway system and river on several discharges.The local scouring phenomenon at its downstream is also taken into account in the analysis.Both the physical and numerical models were used to verify to fulfill dam safety requirements from a hydraulic perspective and to find recommendations.

Donggala in Central Sulawesi
Province is an area with fairly dry climate conditions, so the Surumana Dam was built, which is located at upstream of the Surumana River.This dam functions to provide irrigation water covering an area of 2,100 ha, ponds covering an area of 500 ha, domestic raw water of 1.20 m 3 /second, industry of 2.30 m 3 /second, and electricity generation of 1.5 MW.The total storage volume of the Surumana Dam is 45.14 million m 3 .This reservoir is intended as a reservoir for water runoff or floods as well as to provide water needs during the dry season so that it can improve community welfare and support the economic growth of the people of Donggala Regency.
The hydraulic model test of Surumana Dam spillway was carried out using an undistorted model scale in several stages of flow hydraulics testing for calibration and verification.The model scale in the research is based on many considerations, including the ability of the pump to deliver QPMF to the model, water availability, space availability for model testing, and the expected accuracy with a maximum relative error of 10%.Based on these considerations, a trial error calculation was carried out so that it was determined using an undistorted scale of 1:60, which gave a relative error of 4.75%.This error meets the accuracy requirements, which is less than 10%, for the determined scale to be acceptable.
It is necessary to pay attention to side spillway type building so that the flood discharge that crosses it does not cause a flow that submerges the weir (submerged flow) in the control channel, so the side channel is made relatively low to the weir.To meet this requirement, the spillway system is planned so that there is a difference in elevation when channeling abnormal flood discharges.The water level upstream and downstream of the regulating weir is not less than 2/3 times the water level above the weir's crest.For hydraulic analysis on the spillway, the continuity equation for a steady flow is used [6] as shown in Eq. 1.
Where: Q = discharge (m 3 /sec) V1 = mean velocity in plane-1 (m/sec) A1 = flow cross sectional area in plane-1 normal to the direction of the flow (m 2 ) V2 = mean velocity in plane-2 (m/sec) A2 = flow cross sectional area in plane-2 normal to the direction of the flow (m 2 ) Cavitation is a phenomena that manifests in environments characterized by significant pressure fluctuations in the presence of high-velocity flows, such as the spillway chute of a dam.The presence of an air bubble layer adjacent to the concrete surface serves as a protective barrier, effectively mitigating the impact of high-pressure jets or shockwaves.This phenomenon occurs as vapor or gas bubbles collapse within the fluid flow [7].Due to its effect on the structural concrete surface of the dam, it is The classification of scour encompasses both local and universal aspects [8].Local scour is a phenomenon that occurs due to the dynamic interplay between fluid flow and hydraulic structures.This phenomenon is particularly observed in the proximity of abutments, bridge piers, or downstream channels that discharge into a river.In contrast, it is seen that general scour exhibits a slower rate compared to local scour.General scour refers to the overall erosion of the channel bed, and it can be attributed to various underlying factors.In certain instances, the phenomenon arises as a consequence of a decrease in the cross-sectional area of the channel, resulting in an increase in velocity.Additionally, it manifests in bends and junctions due to the redistribution of flow caused by helical flow and variations in velocity.The extent of erosion occurring downstream of the structure is influenced by its geometric properties, along with the hydrodynamic circumstances and physical attributes of the bed material in that area [8].The estimation of the maximum scour is considered the most pertinent parameter.
The spillway hydraulic model was carried out according to the Q100, Q1000, and QPMF discharge by measuring water level, speed, and pressure height and testing scour patterns.The physical Surumana dam model is shown in Figure 1 to Figure 4. Figure 1 shows the upstream direction channel and spillway, including its sections.The physical model of the transition channel and slide channel are shown in Figure 2, while the energy dissipator and escape channel are shown in Figure 3. Figure 4 shows the model of fixed bed and movable bed in the downstream.

Results and Discussion
Based on the physical model under 1:60 scale for the spillway system, the numerical model is also used to verify the results to understand in several locations where the results of the physical model are difficult to measure precisely.Analysis of the test results of the test model used was at the Q100, Q1000, and QPMF discharge tests for both flow hydraulic conditions and local scour.So, the test results of the Surumana Dam spillway model test based on the hydraulic criteria of the seven parts of the spillway studied from upstream to downstream of the spillway have met the hydraulic rules or criteria for the Q1000 test discharge.
Several points show that the hydraulic criteria are in a condition that meets the recommendations.This happens because field conditions are optimal, so design modifications (design alternatives) can be carried out for the upstream guide channel and launch channel.Flow conditions that show the water velocity are more than the predetermined hydraulic criteria cause the flow to be in a supercritical state and cavitation to occur potentially.Therefore, when it is no longer possible to modify the design (alternative design), it is recommended to add an aerator [7].In this case, the aerator allows air to enter the bottom of the channel, so the pressure at the bottom can be increased.This can prevent the potential for cavitation due to high water speed and pressure.
From the result of the analysis, it was shown that the hydraulic behavior of all parts in the spillway system is capable of channeling all design flood discharges.It was also shown that velocity at the approach channel is less than or equal to 4 m/sec to avoid helicoidal flow by having an aerator [9] and depth of the channel bottom P ≥ (H/5) and subcritical flow conditions.
The side channel is lower than the weir, which is not less than 2/3 of the water level above the spillway.The bottom width of the side channels increases to the downstream with base slope I ≤ (1/13), and it has subcritical flow condition.Furthermore, the transition channel is expected to be in subcritical flow, and this condition is reached by adding an aerator.The critical flow at the hydraulic control point at the downstream end of the transition channel and cavitation negative pressure < -2.33 kPa or nearly -0.3 kg/cm 2 meet its requirements [10].The test results show that the water level is still uneven in the upstream direction channel, and there is sinking flow in the QPMF, but these conditions are adequate for Q1000.It is also found that in the energy dissipator at QPMF flows discharge, a little backwater occurs when the flow waves from downstream are high and do not occur continuously.Adding gabions downstream of the escape channel and downstream of the cliff was proven to give better results to the scour pattern tests.The scour point was reduced, and the depth of the scour was not as deep.
In modeling, scour downstream of the spillway with the 2D numerical model by meshing the model which is divided into active mesh (the area to be observed) and inactive mesh (outside the observed area).The type of basic channel material for the moving part (movable bed) and the non-moving part (fixed bed) are first determined.In this numerical model, the channel roughness's general value is 0.025.Referring to physical model testing, the grain gradation of the river bed material is 0.4 -0.5 mm.
The numerical analysis results for Q100 discharge show a trend of scouring at the transition of the energy dissipator and the river downstream.The maximum scour depth is around 6 meters, located at the edge of the energy absorber/riprap floor boundary with the natural river bed. Figure 5 shows the flow pattern represented by arrows, showing that the flow coming out of the energy dissipator will spread around the downstream of the dam embankment/dam foot.The flow vortices that occur also have a turbulence effect and increase scouring around the downstream dam foot embankments and energyabsorbing buildings.The maximum scour depth in the numerical model for the Q100 discharge during 3 hours of flow is approximately 6 meters, with the maximum scour location as shown in Figure 5. Physical model test results confirm this maximum scour depth as 4.82 meters deep in Figure 6.

Conclusion
The present study investigates the hydraulic behavior characteristic from the upstream of spillway system and river on several discharges (as Q100, Q1000 and QPMF) and its local scouring phenomenon at the downstream.The test results show that in the upstream direction channel, the water level is still uneven and there is sinking flow in the QPMF, but these conditions are adequate for Q1000th.It is also found that in the energy dissipator at QPMF flows discharge, a little backwater occurs when the flow waves from downstream are high and do not occur continuously.The addition of gabions downstream of the escape channel and downstream of the cliff was proven to give better results to the scour pattern tests.The scour point was reduced, and the depth of the scour was not as deep.

Figure 1 .
Figure 1.Physical model upstream direction channel and spillway

Figure 2 . 4 Figure 3 .
Figure 2. Physical model of transition channel and slide channel

Figure 4 .
Figure 4. Physical model of fixed bed and movable bed in the downstream

Figure 5 .Figure 6 .Figure 7
Figure 5.The flow pattern, velocity vector and downstream local scouring under Q100 in numerical model

Figure 7 .Figure 8 .Figure 9 .
Figure 7.The flow pattern, velocity vector and downstream local scouring under Q1000 in numerical model

8 Figure 10 .
Figure 10.Local scouring pattern at the downstream after model test running in QPMF